Probabilistic Forecasting of Electricity Demand Incorporating Mobility Data
Source abstract
Abstract
Due to extreme weather conditions and anomalous events such as the COVID-19 pandemic, utilities and grid operators worldwide face unprecedented challenges. These unanticipated changes in trends introduce new uncertainties in conventional short-term electricity demand forecasting (EDF) since its result depends on recent usage as an input variable. In order to quantify the uncertainty of EDF effectively, this paper proposes a comprehensive probabilistic EFD method based on Gaussian process regression (GPR) and kernel density estimation (KDE). GPR is a non-parametric method based on Bayesian theory, which can handle the uncertainties in EDF using limited data. Mobility data is incorporated to manage uncertainty and pattern changes and increase forecasting model scalability. This study first performs a correlation study for feature selection that comprises weather, renewable and non-renewable energy, and mobility data. Then, different kernel functions of GPR are compared, and the optimal function is recommended for real applications. Finally, real data are used to validate the effectiveness of the proposed model and are elaborated with three scenarios. Comparison results with other conventional adopted methods show that the proposed method can achieve high forecasting accuracy with a minimum quantity of data while addressing forecasting uncertainty, thus improving decision-making.
Source: crossref · Quality: baseline
Research context
OpenAlex topics: Energy Load and Power Forecasting · Integrated Energy Systems Optimization · Electric Power System Optimization
SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.
Affiliations
School of Electrical and Data Engineering, University of Technology Sydney, Ultimo, NSW 2007, Australia
References (66)
Show source-provided references
1. Fatema, I., Kong, X., and Fang, G. (2020, January 28–30). Analyzing and forecasting electricity demand and price using deep learning model during the COVID-19 pandemic. Proceedings of the Parallel Architectures, Algorithms and Programming: 11th International Symposium, PAAP, Proceedings 11, Shenzhen, China.10.1007/978-981-16-0010-4_11 ↗
2. Power system load forecasting using mobility optimization and multi-task learning in COVID-1910.1016/j.apenergy.2021.118303 ↗
3. (2022, August 20). Aggregated Demand and Price Data. Available online: https://aemo.com.au/.
4. (2023, January 20). Australian Energy Market Operator (AEMO 2022). Available online: https://aemo.com.au/.
5. Chen, Y., Yang, W., and Zhang, B. (2006). Using mobility for electrical load forecasting during the covid-19 pandemic. arXiv.
6. Electricity demand and price forecasting model for sustainable smart grid using comprehensive long short term memory10.1080/19397038.2021.1951882 ↗
7. Probabilistic electric load forecasting: A tutorial review10.1016/j.ijforecast.2015.11.011 ↗
8. Review on probabilistic forecasting of photovoltaic power production and electricity consumption10.1016/j.rser.2017.05.212 ↗
9. Hor, C.-L., Watson, S., and Majithia, S. (2006, January 11–15). Daily load forecasting and maximum demand estimation using ARIMA and GARCH. Proceedings of the 2006 International Conference on Probabilistic Methods Applied to Power Systems, Stockholm, Sweden.10.1109/PMAPS.2006.360237 ↗
10. Bozkurt, Ö.Ö., Biricik, G., and Tayşi, Z. (2017). Artificial neural network and SARIMA based models for power load forecasting in Turkish electricity market. PLoS ONE, 12.10.1371/journal.pone.0175915 ↗
11. Load Forecasting Using Support Vector Machines: A Study on EUNITE Competition 200110.1109/TPWRS.2004.835679 ↗
12. Electricity Demand Forecasting by Multi-Task Learning10.1109/TSG.2016.2555788 ↗
13. Neural Network-Based Model Design for Short-Term Load Forecast in Distribution Systems10.1109/TPWRS.2015.2390132 ↗
14. Power Market Load Forecasting on Neural Network With Beneficial Correlated Regularization10.1109/TII.2017.2789297 ↗
15. Wind Power Prediction by a New Forecast Engine Composed of Modified Hybrid Neural Network and Enhanced Particle Swarm Optimization10.1109/TSTE.2011.2114680 ↗
16. GEFCom2012 hierarchical load forecasting: Gradient boosting machines and Gaussian processes10.1016/j.ijforecast.2013.07.002 ↗
17. Hybrid Load Forecasting Method With Analysis of Temperature Sensitivities10.1109/TPWRS.2006.873099 ↗
18. Combining Probability Density Forecasts for Power Electrical Loads10.1109/TSG.2019.2942024 ↗
19. Long Term Probabilistic Load Forecasting and Normalization With Hourly Information10.1109/TSG.2013.2274373 ↗
20. Hyndman, R., Koehler, A.B., Ord, J.K., and Snyder, R.D. (2008). Forecasting with Exponential Smoothing: The State Space Approach, Springer Science & Business Media.10.1007/978-3-540-71918-2 ↗
21. Demand forecasting in power distribution systems using nonparametric probability density estimation10.1109/59.801873 ↗
22. A Bayesian-Based Approach for a Short-Term Steady-State Forecast of a Smart Grid10.1109/TSG.2012.2231441 ↗
23. Probabilistic Load Forecasting via Quantile Regression Averaging on Sister Forecasts
24. On Normality Assumption in Residual Simulation for Probabilistic Load Forecasting10.1109/TSG.2015.2447007 ↗
25. GEFCom2014 probabilistic electric load forecasting: An integrated solution with forecast combination and residual simulation10.1016/j.ijforecast.2015.11.005 ↗
26. Rasmussen, C.E., and Williams, C.K.I. (2008). Gaussian Processes for Machine Learning, MIT Press.
27. Blum, M., and Riedmiller, M. (2013, January 14–18). Electricity demand forecasting using Gaussian processes. Proceedings of the Workshops at the Twenty-Seventh AAAI Conference on Artificial Intelligence, Bellevue, WA, USA.
28. An Integrated Gaussian Process Modeling Framework for Residential Load Prediction10.1109/TPWRS.2018.2851929 ↗
29. Gaussian mixture model coupled recurrent neural networks for wind speed interval forecast10.1016/j.enconman.2019.06.083 ↗
30. Solar radiation intensity probabilistic forecasting based on K-means time series clustering and Gaussian process regression10.1109/ACCESS.2021.3077475 ↗
31. Robust Deep Gaussian Process-Based Probabilistic Electrical Load Forecasting Against Anomalous Events10.1109/TII.2021.3081531 ↗
32. Deep Bayesian Gaussian processes for uncertainty estimation in electronic health records10.1038/s41598-021-00144-6 ↗
33. Laradji, I.H., Schmidt, M., Pavlovic, V., and Kim, M. (2019, January 14). Efficient deep Gaussian process models for variable-sized inputs. Proceedings of the 2019 International Joint Conference on Neural Networks (IJCNN), Budapest, Hungary.10.1109/IJCNN.2019.8851768 ↗
34. Power load probability density forecasting using Gaussian process quantile regression10.1016/j.apenergy.2017.11.035 ↗
35. Short-Term Solar Power Forecasting Based on Weighted Gaussian Process Regression10.1109/TIE.2017.2714127 ↗
36. Residential probabilistic load forecasting: A method using Gaussian process designed for electric load data10.1016/j.apenergy.2018.02.165 ↗
37. Zhang, L., Xie, L., Han, Q., Wang, Z., and Huang, C. (2020). Probability Density Forecasting of Wind Speed Based on Quantile Regression and Kernel Density Estimation. Energies, 13.10.3390/en13226125 ↗
38. Quantile regression: A short story on how and why10.1177/1471082X18759142 ↗
39. Short-term power load probability density forecasting based on Yeo-Johnson transformation quantile regression and Gaussian kernel function10.1016/j.energy.2018.04.072 ↗
40. Alamaniotis, M., Chatzidakis, S., and Tsoukalas, L. (2014, January 2–5). Monthly load forecasting using kernel based gaussian process regression. Proceedings of the 9th Mediterranean Conference on Power Generation, Transmission, Distribution, and Energy Conversion: MEDPOWER, Athens, Greece.10.1049/cp.2014.1693 ↗
41. Leith, D.J., Heidl, M., and Ringwood, J.V. (2004, January 12–16). Gaussian process prior models for electrical load forecasting. Proceedings of the 2004 International Conference on Probabilistic Methods Applied to Power Systems, Ames, IA, USA.
42. Mori, H., and Ohmi, M. (2005, January 6–10). Probabilistic short-term load forecasting with Gaussian processes. Proceedings of the 13th International Conference on, Intelligent Systems Application to Power Systems, Arlington, VA, USA.
43. Towards smart energy systems: Application of kernel machine regression for medium term electricity load forecasting10.1186/s40064-016-1665-z ↗
44. Short-term load forecasting using a Gaussian process model: The influence of a derivative term in the input regressor10.3233/IDT-2012-0143 ↗
45. A hybrid model approach for forecasting future residential electricity consumption10.1016/j.enbuild.2015.09.033 ↗
46. Incorporating Practice Theory in Sub-Profile Models for Short Term Aggregated Residential Load Forecasting10.1109/TSG.2015.2493205 ↗
47. Very Short-Term Nonparametric Probabilistic Forecasting of Renewable Energy Generation—With Application to Solar Energy10.1109/TPWRS.2015.2502423 ↗
48. Gibbs, M.N. (1998). Bayesian Gaussian Processes for Regression and Classification, Citeseer.
49. Kersting, K., Plagemann, C., Pfaff, P., and Burgard, W. (2007, January 20). Most likely heteroscedastic Gaussian process regression. Proceedings of the 24th International Conference on Machine Learning, Corvalis, OR, USA.10.1145/1273496.1273546 ↗
50. Tolvanen, V., Jylänki, P., and Vehtari, A. (2014, January 21–24). Expectation propagation for nonstationary heteroscedastic Gaussian process regression. In Proceedings of 2014 IEEE International Workshop on Machine Learning for Signal Processing (MLSP), Reims, France.10.1109/MLSP.2014.6958906 ↗
51. One-day-ahead probabilistic wind speed forecast based on optimized numerical weather prediction data10.1016/j.enconman.2018.03.030 ↗
52. Thompson, J.R., and Tapia, R.A. (1990). Non-Parametric Function Estimation, Modeling, and Simulation, SIAM.10.1137/1.9781611971712 ↗
53. Bandwidth selection for kernel conditional density estimation10.1016/S0167-9473(00)00046-3 ↗
54. Scott, D.W. (2015). Multivariate Density Estimation: Theory, Practice, and Visualization, John Wiley & Sons.10.1002/9781118575574 ↗
55. Viviani, E., Di Persio, L., and Ehrhardt, M. (2021). Energy Markets Forecasting. From Inferential Statistics to Machine Learning: The German Case. Energies, 14.10.3390/en14020364 ↗
56. Kuo, P.-H., and Huang, C.-J. (2018). An Electricity Price Forecasting Model by Hybrid Structured Deep Neural Networks. Sustainability, 10.10.3390/su10041280 ↗
57. A novel soft computing model (Gaussian process regression with K-fold cross validation) for daily and monthly solar radiation forecasting (Part: I)10.1016/j.renene.2017.08.061 ↗
58. Enhanced performance Gaussian process regression for probabilistic short-term solar output forecast10.1016/j.ijepes.2021.106916 ↗
59. Point and prediction interval estimation for electricity markets with machine learning techniques and wavelet transforms10.1016/j.neucom.2013.02.039 ↗
60. Probabilistic forecasts, calibration and sharpness10.1111/j.1467-9868.2007.00587.x ↗
61. (2022, September 26). National Electricity Market. Available online: https://opennem.org.au/energy/nem/.
62. Google (2022, October 15). COVID-19 Community Mobility Reports 2020. Available online: https://www.google.com/covid19/mobility/.
63. Effective use of Spearman’s and Kendall’s correlation coefficients for association between two measured traits10.1016/j.anbehav.2015.01.010 ↗
64. Predicting electricity energy consumption: A comparison of regression analysis, decision tree and neural networks10.1016/j.energy.2006.11.010 ↗
65. Decomposition of the continuous ranked probability score for ensemble prediction systems10.1175/1520-0434(2000)015<0559:DOTCRP>2.0.CO;2 ↗
66. (2022, December 10). NSW State Emergency Service, Available online: https://www.ses.nsw.gov.au/.